Combined MRI, high-resolution manometry and a randomised trial of bisacodyl versus hyoscine show the significance of an enlarged colon in constipation: the RECLAIM study

Abstract

Background: Colonic motility in constipation can be assessed non-invasively using MRI.

Objective: To compare MRI with high-resolution colonic manometry (HRCM) for predicting treatment response.

Design: Part 1: 44 healthy volunteers (HVs), 43 patients with irritable bowel syndrome with constipation (IBS-C) and 37 with functional constipation (FC) completed stool diaries and questionnaires and underwent oral macrogol (500-1000 mL) challenge. Whole gut transit time (WGTT), segmental colonic volumes (CV), MRI-derived Motility Index and chyme movement by 'tagging' were assessed using MRI and time to defecation after macrogol recorded. Left colonic HRCM was recorded before and after a 700 kcal meal. Patients then proceeded to Part 2: a randomised cross-over study of 10-days bisacodyl 10 mg daily versus hyoscine 20 mg three times per day, assessing daily pain and constipation.

Results: Part 1: Total CVs median (range) were significantly greater in IBS-C (776 (595-1033)) and FC (802 (633-951)) vs HV (645 (467-780)), p<0.001. Patients also had longer WGTT and delayed evacuation after macrogol. IBS-C patients showed significantly reduced tagging index and less propagated pressure wave (PPW) activity during HRCM versus HV. Compared with FC, IBS-C patients were more anxious and reported more pain. Abnormally large colons predicted significantly delayed evacuation after macrogol challenge (p<0.02), impaired manometric meal response and reduced pain with bisacodyl (p<0.05).Part 2: Bisacodyl compared with hyoscine increased bowel movements but caused more pain in both groups (p<0.03).

Conclusion: An abnormally large colon is an important feature in constipation which predicts impaired manometric response to feeding and treatment responses. HRCM shows that IBS-C patients have reduced PPW activity.

Trial registration number: The study was preregistered on ClinicalTrials.gov, Reference: NCT03226145.

Keywords: COLORECTAL MOTILITY; COLORECTAL PHYSIOLOGY; CONSTIPATION; LAXATIVES.

Figure 1. MRI Motility Index (MMI), ascending Colon Content Movement and Total Colonic Volumes. (A) Ascending colon MMI This rose significantly over time (p<0.001) ANOVA showed effect of time p<0.001, effect of group NS, p=0.97. (B) Ascending colon content motion was assessed by Tagging index at baseline and 60 (T60) and 120 min (T120) after macrogol ingestion. Tagging index showed a significant increase over time, which was less than HV in IBS-C (*p=0.02) but not in FC (p=0.08) at 120 min (two-way ANOVA, Tukey’s MC.) (C) Total colonic volumes. These rose over time for all groups. Both FC and IBS-C total colonic volumes were greater than HVs but not different from each other two-way ANOVA, Time effect p<0.0001, group effect p=0.0019, post hoc comparisons using Tukey’s multiple comparisons *p<0.05, **p<0.01, ***p<0.001 vs HV. ANOVA, analysis of variance; FC, functional constipation; HV, healthy volunteer; IBS-C, irritable bowel syndrome with constipation; NS, not significant.

Figure 2. Pain on MRI study day pain score (0–3) is shown at baseline, 60 (T60) and 120 (T120) minutes after macrogol^R ingestion. IBS-C and FC had significantly more pain than HVs, p<0.05 at all 3 time points and at T60 IBS-C>FC, p<0.05. Mixed effect model (Restricted maximum likelihood) with Tukey’s MC. FC, functional constipation; HVs, healthy volunteers; IBS-C, irritable bowel syndrome with constipation.

Figure 3. High-resolution colonic manometry before and after meal in HV, IBS-C and FC. (A) Cyclic motor patterns (CMPs) in the sigmoid colon after meal. The percentage of time occupied by the CMPs in the sigmoid colon 1 hour after the meal was significantly lower in IBS-C (27 (SD 19)%) but not FC group (38 (SD 18)%) compared with HVs (38 (18)), * = p=0.049, ANOVA with Tukey’s MCs. (B) High-resolution manometry summary data for propagated PWs in HV versus IBS-C and FC. This shows premeal and postmeal frequency distribution on the vertical axis and phase on the horizontal axis. Points to the right of 0 indicate antegrade propagated waves while those to the left indicate retrograde. Higher power is indicated by yellow showing IBS patients had significantly less power than HV or FC, both premeal and postmeal (for full analysis see online supplemental D2 and G2). ANOVA, analysis of variance; FC, functional constipation; HV, healthy volunteer; IBS-C, irritable bowel syndrome with constipation; PW, pressure wave.

Figure 4. One-dimensional (1D) analysis of pressure waves (PW) at frequencies between 1 and 8 cycles per minute (cpm) in the sigmoid colon. Patients with a normal volume colon (top row; a–c) and patients with an enlarged volume colon (middle row; d–f) left column (a, d, g) shows baseline and middle column (b, e, h) the meal periods. In each image, the frequency of PW is shown on the y-axis. In (a, b, d, e) power is shown on the x-axis. The power refers to the prevalence of the PW at any of the calculated frequencies. 2000 overlapping grey lines in each panel represent posterior samples, and the dotted black lines form envelopes of 95% credible intervals. (g, h) The power ratio across the frequency range, between the enlarged and normal colons. When the entire envelope lies to one side of the vertical red line (which represents a ratio of 1), this shows a significant deviation (to the left a decrease in PWs in the enlarge compared with normal colons; to the right of the red line indicates an increase in PWs). During baseline and a meal period the red lines in (g, h) lie entirely within the grey envelope indicating no significant difference. (c, f) The ratio of the power of baseline activity to meal activity. In the normal colons (c), the grey envelope lies to the right of the red line indicating that the meal induced a significant increase in power in frequencies between 1 and 8 cpm. In patients with an enlarged colon (f), no meal response is seen (red line lies within the grey envelope). (i) The meal effect between the two groups, as the grey envelope lies to the left of the red line indicates that patients with an enlarged colon have a significantly reduced meal response compared with those with a normal diameter colon.

Figure 5. Two-dimensional (2D) analysis of propagating pressure waves (PPW) in the sigmoid colon at frequencies between 1 and 8 cpm. In each panel, the vertical line at 0 on the x-axis indicates synchronous (non-propagating) activity. Retrograde propagation is to the left of the midline and antegrade to the right. The curved dotted lines indicate the speed of propagation, from 3 cm/min to 100 cm/min. (a, b, d, e) The green pixels represent the increasing power of propagated activity. The first column represents baseline data, the second column meal data. Patients with a normal diameter colon are shown in the top row, patients with an enlarged colon in the second row. The bottom row compares PPW power across the frequency range between the normal and enlarged colon during baseline (g) and meal (h) periods. (g) The orange area demarcated by the solid white line indicates a significant increase in antegrade and retrograde PPW at <3 cpm in the enlarged compared with normal volume colons. (h) The blue area demarcated by the solid white line indicates a significant decrease in antegrade and retrograde PPW between 2 and 4 cpm in patients with enlarged compared with normal volume.